Download Web Mining and Data Mining

International Journal of Novel Research in Computer Science and Software Engineering
Vol. 2, Issue 1, pp: (36-42), Month: January - April 2015, Available at: www.noveltyjournals.com
Web Mining and Data Mining: A Comparative
Approach
1
1,2
Simranjeet Kaur, 2Kiranbir Kaur
Department of Computer Science & Engineering, Guru Nanak Dev University, Amritsar, India
Abstract: Search engine is used to retrieve data from the web in response to user’s queries. Web mining is a
technique of data mining, which is used to discover data from the web documents. With the help of Web mining,
the user obtains the required information accurately. Web mining is categorized into three types: web content
mining, web structure mining and web usage mining. In this paper, difference of three categories of web mining
and comparison of web mining and data mining is also presented. Data mining helps to extract data from
database, whereas data is extracted from web in web mining.
Keywords: Web Mining, Data Mining, Web Structure Mining, Web Content Mining, Web Usage Mining.
I.
INTRODUCTION
As the World Wide Web is very enormous and contain all sort of information. With the growing information needs and
sources, it becomes necessary to manage the information and to display most relevant information to the user screen
through search engine.
Users extract information from www with the use of search engine as information retrieval tool. Commonly used search
engines are Google, Yahoo, Bing, etc. They index, download and store billions of web pages, and act as content
aggregators because they keep every record of web pages available on the WWW [1].
Figure 1: Concept of search engine [1]
There are primary two kinds of search services available on the internet [2]:
A. Organic Search Engine: To index websites into large database, automated programs known as “spiders” or “robots”
are used. Spider index entire website when user submits page to organic search engine. Examples are Google, MSN,
Yahoo, etc.
B. Search Directories: It include list of categories reviewed by human that rely on site owner’s submission. It’s all
reviewer responsibility to describe the site in what text. Examples include Open Directory (Google Directory), Yahoo
Directory, Look smart and others.
Page | 36
Novelty Journals
International Journal of Novel Research in Computer Science and Software Engineering
Vol. 2, Issue 1, pp: (36-42), Month: January - April 2015, Available at: www.noveltyjournals.com
II.
OVERVIEW OF SEARCH ENGINE OPTIMIZATION
Search Engine Optimization (SEO) technique is a part of Search Engine Marketing (SEM), to increase the traffic to
websites. It helps the spider program to find the relevant webpage and provide it higher rank on the web, by analyzing the
webpage content, keywords and webpage code [3].
III.
EVOLUTION OF WEB MINING
Web Mining process began when the business data started storing on computers and internet and continued with
improvements in data access, and with use of WWW. Data is collected from various sources like surveys, networked
locations through the use of computers, and used as required [7].
Web mining is a revolutionary process, and the user gets the answer quickly and accurately. The evolution of web mining
is shown below in table 1:
Table I: Evolution of web mining [7]
Evolutionary Steps
1960s
Data Collection
1980s
Data Access
1990
Data Warehousing
and Decision
Support
Technologies
Product Providers
Business Need
Characteristics
Computers, disks
CDC, IBM
To know total
Delivery of static
and tapes
revenue in last years data
SQL, RDBMS,
Oracle, IBM ,
To know about unit
Dynamic data
ODBC
Sybase, Microsoft
sales
delivery
Multi-dimensional
Microstrategy,
To know about unit
Dynamic data
database, data
Arbor, Congnos,
sale and also to take
delivery at multiple
warehouses, OnPilot
decision on it
levels
Line Analytic
Processing
Multiprocessor
IBM, SGI, Pilot,
What to do with data Prospective
2000s
computers, advanced Lockheed
and why
information delivery
Data Mining
algorithms, Massive
databases
WWW, Internet,
IBM, Web Trends,
To know about
Affordable tool to
Emerging Today
monumental scale
Net Genesis,
business in past or
mine large data
Web Mining
database, learning
Rockware, Apecto
future
warehouse and
technology like
Limited, Heckyl
relational databases
supervised like rule
Technologies
efficiently and fast
generalization and
using multiple
unsupervised like
mining functions,
pattern mining
Powerful
Web Mining is a data mining technique that helps to extract or discovers information from the web documents. Web data
can be in the form of web pages like text and images, inter-page structure i.e. linkage structure between web pages, intra
page structure i.e. include XML or HTML tags and about the usage of data on the internet [5].
IV.
WEB MINING PROCESS
The process of extracting knowledge from web is as follows [4]:
Figure 2: Web Mining Process [4]
Page | 37
Novelty Journals
International Journal of Novel Research in Computer Science and Software Engineering
Vol. 2, Issue 1, pp: (36-42), Month: January - April 2015, Available at: www.noveltyjournals.com
In Web mining, data mining is used to extract the hidden data in web log (usage data). The subtasks of web mining are
shown in figure 3:
Figure 3: Subtasks of web mining [7]
V.
WEB MINING CATEGORIES
Web Mining is categorized into three types: web content mining, web structure mining and web usage mining as shown in
figure 4.
Figure 4: Categories of Web Mining [7,9]
A. Web Content Mining
It is the process to retrieve information from web into structured form and index the information, and extract it quickly. Its
main focus is on the structure on the inner document and it consists of text, images, audio, video or structured records
such as tables and lists [6].
Page | 38
Novelty Journals
International Journal of Novel Research in Computer Science and Software Engineering
Vol. 2, Issue 1, pp: (36-42), Month: January - April 2015, Available at: www.noveltyjournals.com
B. Web Structure Mining
It is a process to model the linking structure of web pages. Its main motive is to generate structured summary of web
pages and websites [6]. This is carried out at the hyperlink level (inter-page) and at document level (intra-page). This type
of mining is useful for retrieving information [4].
Figure 5: Web graph structure [8]
Web structure mining consists of three categories [9]:

Link Structure: It include link based cluster analysis, link based classification, link type and link strength.
 Internal Structure Mining: It provides the information about ranking of pages, and to enhance the search results by
discovering the model underlying the link structure. It is also helpful to analyze the relationship between different
websites.
 URL Mining: In this category, hyperlink is used to connect a web page to different locations, either within same or
different web page.
C. Web Usage Mining:
This type of mining is used to discover useful information and navigation patterns from the web present in server logs,
agent logs, referrers log, client-side cookies, Meta data and user profile. It introduces privacy concern [8].
Figure 6: Web usage mining process [8]
When the user interacts with web, it is used to predict user behavior [7]. It consists of three phases:
 Preprocessing: Proxy server and server is the first way used to retrieve raw data from the web and process it, and it
makes automatic transformation of original raw data [9].
 Pattern Discovery: After discovering the knowledge from preprocessing, techniques are implemented to discover the
knowledge such as machine learning, data mining procedures and other related procedures [9].
 Pattern Analysis: After the stage of pattern discovery, pattern analysis work is to check the pattern on the web and
its implementation to extract the knowledge from the web [9].
Page | 39
Novelty Journals
International Journal of Novel Research in Computer Science and Software Engineering
Vol. 2, Issue 1, pp: (36-42), Month: January - April 2015, Available at: www.noveltyjournals.com
Table II: Categories of Web Mining [6, 9]
Web Mining Categories
Specification
View of data
Web Content Mining
Structured
Semi-structured
Unstructured
Primary
Text document
Hypertext document
Concepts or ontology
Relational
Edge labeled graph
n-grams
Terms, Phrases
Machine learning
Association rules
Proprietary algorithm
Statistical method
It tells about the discovery
of useful information from
web documents/contents
Web Structure Mining
Structure linkage
Web Usage Mining
Interactive data
Primary
Link structure
Secondary
Browser logs
Server logs
Relational table
Graph
Scope
Type of data used
Main data
Representation
Graph
Proprietary algorithm
Machine learning
Statistical method
It discover the model
underlying web’s link
structure
Its work is to make sense
of the data generated by
web surfer’s behavior
The scope of data is local in
DB and is global in IR.
Global
Global
Goal
It mainly targets to discover
knowledge
To generate structural
summary about the web
page and web site [10]
To analyze the behavioral
pattern and users profile
while interacting with web
sites
Application areas
Clustering
Categorization
Finding extraction rules and
patterns
User modeling
Finding frequent subschemas
Data/Information extraction
Opinion extraction from
online sources
Segmenting web pages and
noise detection [10]
Clustering
Categorization
User modeling
Site construction
Marketing
Adaptation and
Management
Not all pages have relevant
meta information
Entire text of predecessor
page is not relevant
Preprocessing challenges
about the users i.e. who
will be user, how long it
stay, where will it go and
user view?
Method
Tasks
Challenges
VI.
COMPARISON OF WEB MINING AND DATA MINING
Data Mining (also called data or knowledge discovery) is the process to find hidden information or patterns in databases
[11], and summarizing it into useful information. It is also the process to find co-relations among large number of fields in
huge relational databases [17], whereas Web Mining is a technique that is used to retrieve the documents from the web.
Data Mining includes the five major steps: extract, store and manage data, provide data access, analyze the data and
present it in a useful form, like in the form of table or graph [17].
Page | 40
Novelty Journals
International Journal of Novel Research in Computer Science and Software Engineering
Vol. 2, Issue 1, pp: (36-42), Month: January - April 2015, Available at: www.noveltyjournals.com
Table III: Comparison of Web Mining and Data Mining [19]
Web Mining VS Data Mining
Comparison
Web Mining
Data Mining
Definition
Process used to extract information
from web documents.
Process used to extract hidden
information from the database.
Scale
It contains 10 million jobs in server
database, and therefore search
processing is not big.
It contains 1 million jobs in database
and search processing is large.
Structure
The information is obtained from
structured, semi-structured and
unstructured web forms. It gets the
information from wide database.
It obtains the information from
explicit structure. It is not able to get
all the information from wide
database as compared to web
mining.
Access
Data is accessed publicly. In this,
data is not hidden in web database
and only permission is required to
access the data from web log master.
Data is accessed privately and only
authorized user can access the data.
Data
It works upon on-line data.
It works upon off-line data.
Data Storage
Data is stored in server logs and web
server database.
Data is stored in data warehouses.
Application Areas
E-learning, Digital Libraries, EGovernment, Electronic Commerce,
E-Politics, E-Democracy, Security &
Crime Investigation, Electronic
Business [13].
Banking, marketing, manufacturing
& production, health-care,
insurance, law, airlines, computer
hardware & software, government &
defence, etc [12].
Disadvantages
url’s can be tracked to access the
data, multiplicity of events and url’s,
large amount of data remain unused
Privacy issues, security issues,
misuse of information/ inaccurate
information [12].
Challenges
Complexity of web pages, web is too
huge, relevancy of information, web
is dynamic information source,
diversity of user communicates, etc
[14].
Network settings, data quality,
privacy preservation, scalability,
complex and heterogeneous data, etc
[12].
Techniques
Web Content Mining, Graph Based
Web Mining, Utilization in Web
Mining, Text Mining and many
others [15].
Artificial Neural Network, Decision
Trees, Rule Induction, Nearest
Neighbor Method and many others
[16].
VII.
CONCLUSION
The need to access and manage data on the web is increasing day by day. Web mining enhances user's ability to access
information from www and finds its applications in various fields like Clustering, Categorization, in extraction rules and
patterns and many others. In the comparison of web mining with data mining, it is concluded that web mining is used to
retrieve online data, and data mining retrieves offline data. Data is stored in server database in web mining and it can
handle multiple transactions at the same time. Data can be discovered and extracted from multiple locations of the world
by sitting at one location and is able to provide desired information at the time of requirement.
Page | 41
Novelty Journals
International Journal of Novel Research in Computer Science and Software Engineering
Vol. 2, Issue 1, pp: (36-42), Month: January - April 2015, Available at: www.noveltyjournals.com
REFERENCES
[1]
N. Duhan, A.K. Sharma and K. K. Bhatia, “Page Ranking Algorithms: A Survey”, IACC 2009
[2]
Clifton and N. Rae, “How Search Engine Optimization Works”, SEO Whitepaper
[3]
Z. Hui, Q. Shigang, L. Jinhua and C. Jianli, “Study on Website Search Engine Optimization”, International
Conference on Computer Science and Service System, 2012
[4]
N. Tyagi and S. Sharma, “Comparative study of various Page Ranking Algorithms in Web Structure Mining”,
IJITEE, ISSN: 2278-3075, Volume-1, Issue-1, June 2012
[5]
Jain, R. Sharma, G. Dixit and V. Tomar,” Page Ranking Algorithms in Web Mining, Limitations of Existing
methods and a New Method for Indexing Web Pages”, International Conference on Communication Systems and
Network Technologies, 2013
[6]
R. Jain and Dr. G.N Purohit, “International Journal of Computer Applications (0975-8887) Volume 13-No.5,
January 2011
[7]
K. Sharma, G. Shrivastava and V. Kumar, “Web Mining: Today and Tomorrow”, IEEE, 2011
[8]
M. T. Ramakrishna, L. K. Gowdar, M. S. Havanur and B. P. M. Swamy, ”Web Mining: Key Accomplishments,
Applications and Future Directions”, International Conference on Data Storage and Data Engineering, 2010
[9]
B. Rathod and Dr. S. Khanna, “A Review on Emerging Trends of Web Mining and It’s Applications”, IJEDR,
ISSN: 2321-9939
[10]
Web Content Mining Problems/Challenges, available at, “https://sites.google.com/site/assignmentssolved /mca/
semester6 /mc0088/12”
[11]
N. Padhy, Dr. P. Mishra and R. Panigrahi, “The survey of Data Mining Applications And Future Scope”,
International Journal of Computer Science, Engineering and Information Technology, Vol.2, No.3, June 2012
[12]
S. H. Begum, “Data Mining Tools and Trends – An Overview”, International Journal of Emerging Research in
Management & Technology, ISSN: 2278-9359, February 2013
[13]
S. Yadav, K. Ahmad and J. Shekar, “Analysis of Web Mining Applications and Beneficial Areas”, IIUM
Engineering Journal, Vol. 12, No. 2, 2011
[14]
Data Mining – Mining World Wide Web available at “http://www.tutorialspoint.com/data _mining/dm_ mining_
www.htm”
[15]
P. Bhisikar and A. Sahu, “Overview on Web Mining and Different Technique for Web Personalization”, In
International Journal of Engineering Research and Applications, ISSN: 2248-9622, Vol. 3, Issue 2, March-April
2013
[16]
An Introduction to Data Mining, available at, “http://www.thearling.com/text/dmwhite/dmwhite.htm”
[17]
Data Mining: What is Data Mining? Available at,”http://www.anderson.ucla.edu/faculty/jason. frand/teacher/
techno -logies/palace/datamining.htm”
[18]
Rastogi, S. Gupta, S. Agarwal, N. Agarwal, ”Web Mining: A Comparative Study”, International Journal of
Computational Engineering Research, ISSN: 2250-3005, Vol. 2, Issue No. 2, Mar-Apr 2012
[19]
R. Sharma, “A Framework to Compare Web Mining Types”, In International Journal of Advanced Research in
Computer Science and Software Engineering, ISSN: 2277-128X, Volume 3, Issue 7, July 2013
Author Profile:
Simranjeet Kaur belongs to Amritsar, Punjab (India) and her Date of Birth is 15 December, 1989. She is B.Tech (IT)
and pursuing M.Tech (Software System) from Guru Nanak Dev University in the Department of Computer Science and
Engineering, Amritsar, India. Her interest area is search engine optimization and web information retrieval. She
completed this paper under the guidance of Kiranbir Kaur, Assistant Professor in Guru Nanak Dev University,
Department of Computer Science and Engineering, Amritsar, India.
Page | 42
Novelty Journals